Commodities in practice

The Dutch industry-wide pension fund for the healthcare and social work sector PGGM - with around €80bn assets under management - was one of the first European investors to invest large amounts when it started commodity investments in 2000. The driver behind investments in commodities was diversification.

Frans de Wit, senior investment manager of PGGM's commodity portfolio, says: "In the late 1990s, PGGM changed its allocation away from fixed income towards other higher-yielding asset classes and with that we needed more diversification within our portfolio.

"If you look at the returns over the years, it was the right time to start investing. It has effectively been the negative to zero correlation with other asset classes that has achieved a lower amount of risk with the same expected return in the overall portfolio."

Other pension funds such as ABP followed PGGM's example shortly afterwards.

In the US, the California Public Employees' Retirement System (CalPERS) is currently in a market watch mode until an assessment on its commodities pilot programme for its board has been made. Investment staff will give the board details of the proposed inflation-linked investment asset class at a July offsite meeting before the key decisions are made at the CalPERS investment committee's asset allocation workshop in November.

The board will decide at the workshop whether it will go forward with commodities as a sub-asset class of the inflation-linked new asset class, which on top of commodities contains infrastructure, timber and inflation-linked bonds.

As a rule, the amounts invested in commodities are small. PGGM currently invests 5%, around $5.5bn, of its total assets, in commodities. The pension fund started off with 4% in 2000 but increased the allocation to the current level in 2005.

In this process, PGGM also changed its benchmark from 100% Goldman Sachs Commodity Index (GSCI) to 75% GSCI plus 25% GSEN sub energy index and then in 2005 to the current benchmark of 60% GSCI index with all its 24 underlying commodities plus 40% in the GSEN sub energy index. This gives PGGM an energy exposure of around 80%.

ABP has a strategic allocation to commodities of 3% as part of its long-term strategic investment plan. This has also slowly risen since its first commodity investment in 2001.

Commodity investment levels are not bound by Dutch law and some studies from 2000 indicate a 15-20% optimum level. De Wit says PGGM started out with 4% due to the limited capacity of the investor market in 2000, which was roughly $6-10bn (€4.5-7.5bn), with PGGM covering a quarter of that.

On top of that, he says, it was a reputational risk due to its derivative nature and because it was a novelty for a pension fund to invest in at the time.

PGGM's commodities exposure is based on the pension fund's asset-liability (ALM) study that is undertaken every three years. And based on this, the 5% the pension fund currently invests is the optimum investment level for PGGM.

 

The GSCI has been the benchmark of choice for many pension fund investors in commodities in the brief history of institutional interest in the asset class. CalPERS has so far committed $450m to the GSCI. The index is an important benchmark to the pension fund, but one that CalPERS would like to outperform while maintaining its risk-reward benefits.

Clark McKinley, a spokesperson for CalPERS, says: "The plus is coming out of our use of swaps, notes and related investments that are rolled on a monthly basis from one contract to another to help us outperform established indexes on the margin."

De Wit adds: "The ALM study decided about our method of commodity investing after the analysis of several indices. The index had to have the specific characteristics that we were looking for within commodities as well as being replicable. In other words, it could not have been a proprietary index or an index that rebalances with every underlying tick.

"In the end, we were left with two indices - the Dow Jones AIG Index and the GSCI. We chose the GSCI because it has a higher energy weighting and our ALM study showed that it is the energy component within commodities that gives the best diversification."

He explains that if it had been up to the current manager at the time, PGGM would have eventually shifted to a full energy benchmark. But being one of the first investors, PGGM initially wanted to err on the side of caution. Since then, however, it has gradually shifted towards a higher-energy weighting.

"The method pension funds select to invest in commodities depends first of all on the reason why they are investing in commodities and their infrastructure," continues de Witt. "If they - like us - invest for the negative correlation and [are] trying to lower the overall amount of risk in the portfolio, it's an ideal approach. However, if they look at it from an alpha approach, I don't think it's the right way."

 

PGGM's investment in commodities brought net returns of 35.5% in 2002 and returns in the early to mid twenties in the subsequent three years, but negative returns of -22.3% in 2006. ABP saw net returns in commodities of 18.8% in 2004 and of 23.2% in 2005 but returns for the third quarter of 2006 were also negative at -15.4%.

When deciding whether to implement commodities, PGGM used prudent metrics with a return similar to that of fixed income and a standard deviation comparable to that of equities and private equity.

"In that sense, a stand-alone investment in commodities does not make sense, but it does if you add the negative to zero correlation with the other assets in the mix, which makes the whole portfolio more robust," says de Witt.

PGGM looks at commodities from a risk-return and complete portfolio perspective and started its alpha-beta approach in early 2006, giving up its separate commodities desk and moving commodities into the overall beta portfolios.

"The risk involved is not so much the volatility - in fact it is the higher volatility that we are looking for as long as the negative correlation works," de Witt adds.

PGGM is in it for the long run, but de Wit says that commodities are periodically revalued as part of the ALM study. He says: "The fact that the negative correlation doesn't always work, and in some years goes against us, really shows that we are in it for the long-term. And looking ahead, the most important reason is the existence of the negative to zero correlation with the other asset classes to determine whether we will still be investing in commodities the way we are now.

"We would probably not give a higher allocation to commodities at the moment because the risk that has freed up is now being used in other areas."

And McKinley says: "Going forward, we believe that up to 50% of our private equity investments could be concentrated in the energy and raw materials sectors. That is a huge shift from zero to 50% and ethanol plants are but one of the opportunities we are evaluating."

The fund has already taken a private equity stake in an ethanol plant, which will make sense in the future according to McKinley, assuming oil trades above $40 a barrel. "So we would sell crude oil or ethanol futures to effectively lock in a profit at our ethanol facilities," explains the CalPERS spokesperson.

"We'd sell the contracts over a period of six years. It's a bona fide hedge because we are essentially long ethanol in the process of developing the facility. Selling ethanol and oil would reduce our market risk. It is not a directional view on commodities."